​
[skop.dev]

​
Overview

Skop is an intelligent multi-agent document discovery and extraction service that can find and extract PDF documents from websites based on natural language prompts. The API uses advanced AI agents to navigate websites, identify relevant documents, and extract them with high accuracy. Base URL: https://api.skop.dev
Version: 1.0.0
Service Name: Skop

​
Authentication

All API endpoints require authentication using either:
  • Bearer Token: Include Authorization: Bearer <your_token> in headers
  • API Key: Include X-API-Key: <your_api_key> in headers

​
Required Scopes

  • scrape:create - Required for creating new scraping jobs
  • Additional scopes may be required for other operations

Get API Key

Visit the Skop dashboard to create an account and get your API key

​
Billing & Pricing

​
Pricing Model

The API uses a results-based billing system where you only pay for pages that contain actual documents. Our LangGraph agent dynamically allocates different tiers of scraping agents:
  • Advanced Scraping Agent: $0.18 per page with results (high-quality extraction)
  • Standard Scraping Agent: $0.10 per page with results (basic extraction)

​
Usage-Based Billing

The API uses a simple pay-per-use model where all users are charged based on actual results:
  • No monthly limits: Use as much as you need
  • No plan restrictions: All features available to all users
  • Fair pricing: Only pay for pages that contain documents
  • Transparent costs: Clear per-page pricing with no hidden fees

​
Cost Estimation

  • Single Page: ~$0.10-0.18 (depending on agent tier allocated)
  • Multi-Page: Variable based on documents found (typically $1-5 per job)

​
Quickstart

​
1. Get Your API Key

Visit https://www.skop.dev/dashboard to create an account and get your API key.

​
2. Make Your First Request

curl -X POST "https://api.skop.dev/scrape" \
  -H "Authorization: Bearer <your_token>" \
  -H "Content-Type: application/json" \
  -d '{
    "website": "https://example.com",
    "prompt": "Find all board meeting minutes from 2024",
    "parameters": {
      "single_page": true,
      "timeout": 1800,
      "confidence_threshold": 0.7
    }
  }'

​
3. Check Job Status

curl -X GET "https://api.skop.dev/scrape/status/{job_id}" \
  -H "Authorization: Bearer <your_token>"

​
4. Get Results

curl -X GET "https://api.skop.dev/scrape/results/{job_id}" \
  -H "Authorization: Bearer <your_token>"

​
Rate Limits

  • General endpoints: 100 requests per hour
  • Job creation: 100 requests per hour
  • Status checks: 300 requests per hour
  • Health checks: 60 requests per minute
Rate limit headers are included in responses:
  • X-Rate-Limit-Remaining: Remaining requests
  • X-Rate-Limit-Reset: Reset timestamp

​
Request & Response Format

All requests and responses use JSON format with Content-Type: application/json.

​
Error Response Format

{
  "error": true,
  "message": "Human-readable error message",
  "status_code": 400,
  "path": "/scrape",
  "timestamp": "2024-01-15T10:00:00Z",
  "details": {
    "field": "website",
    "reason": "URL must use http or https protocol"
  }
}

​
Common Error Codes

HTTP StatusError TypeDescription
400Bad RequestInvalid request format or parameters
401UnauthorizedMissing or invalid authentication
402Payment RequiredInsufficient credits or billing issue
403ForbiddenAccess denied or feature not available
404Not FoundResource not found
429Too Many RequestsRate limit exceeded
500Internal Server ErrorUnexpected server error
503Service UnavailableService temporarily unavailable

​
Job Status Values

StatusDescription
pendingJob queued, waiting to start
in_progressJob is actively running
completedJob finished successfully
failedJob failed due to errors
cancelledJob was cancelled by user